Novel fingerprint image enhancement algorithm using phase preserving
一种新的基于相位保持的指纹图像增强算法

Phase information of an image plays a very important role during perceiving image by human eyes, and it will help to keep fingerprint texture characteristics unchanged if phase corruption of image can be avoided in a fingerprint image enhancement algorithm. Firstly, an experiment was designed to show that a significant amount of the defining information of fingerprint image texture was contained in phase information. Then a novel fingerprint image enhancement algorithm based on phase preserving was proposed. In the algorithm, complex Gabor wavelets were adopted to extract phase information in fingerprint image, and local orientation information was utilized to enhance fingerprint image under the principle of phase preserving. To evaluate the performance of the algorithm, two kinds of low quality fingerprint images were enhanced. Experiment results show that the quality of enhanced images can be improved significantly.